US20050249376A1 - Mail recognition method - Google Patents
Mail recognition method Download PDFInfo
- Publication number
- US20050249376A1 US20050249376A1 US11/122,241 US12224105A US2005249376A1 US 20050249376 A1 US20050249376 A1 US 20050249376A1 US 12224105 A US12224105 A US 12224105A US 2005249376 A1 US2005249376 A1 US 2005249376A1
- Authority
- US
- United States
- Prior art keywords
- character recognition
- recognition process
- data
- images
- block
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C3/00—Sorting according to destination
- B07C3/10—Apparatus characterised by the means used for detection ofthe destination
- B07C3/14—Apparatus characterised by the means used for detection ofthe destination using light-responsive detecting means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/98—Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
- G06V10/987—Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns with the intervention of an operator
Definitions
- the present invention relates to a mail recognition method.
- Mail recognition methods are known in which images of mail items undergo a first character recognition process to generate a first set of data representing alphanumeric characters (indicating the postal code, receiver's address and data) on the mail items.
- the first recognition process is normally an OCR process, and usually succeeds in recognizing the address on the basis of the alphanumeric characters (roughly 10% of the images fail to be recognized, and the recognition error margin of the OCR process is about 2-3%).
- the images the first recognition process has failed to recognize i.e. about 10%
- the second recognition process may, for example, be an RCR (Reject Character Recognition) process, which usually succeeds in recognizing 40% of the preceding OCR process rejects.
- RCR Reject Character Recognition
- recognition algorithms may be variously parameterized to improve recognition probability at the expense of recognition accuracy (probability complementary to error probability per recognized volume).
- recognition and accuracy percentages may be insufficient to ensure successful follow-up mail processing, such as highly automated sorting processes.
- a mail recognition method comprising the steps of: acquiring images of mail items, and subjecting said images to a first character recognition process to recognize addresses; and characterized by comprising the step of subjecting the images successfully processed by said first character recognition process to a second character recognition process, different from the first, to recognize an address; and checking to determine whether the results of the different character recognition processes performed on the same image match.
- FIG. 1 shows, schematically, a mail recognition system implementing the method according to the present invention
- FIG. 2 shows first steps in the method according to the present invention
- FIG. 3 shows second steps in the method according to the present invention
- FIG. 4 shows third steps in the method according to the present invention
- FIG. 5 shows a table used in the method according to the present invention.
- Number 1 in FIG. 1 indicates as a whole a mail processing system implementing the method according to the present invention.
- system 1 comprises:
- the second read, recognition and sorting unit 9 emits mail items 7 (letters, postcards, enclosed documents, flats, cellophane-wrapped magazines, unwrapped magazines, catalogues, etc.), and possibly also sends them to other units (not shown) for performing known mail processing operations, such as a mail sequencing process.
- mail items 7 letters, postcards, enclosed documents, flats, cellophane-wrapped magazines, unwrapped magazines, catalogues, etc.
- the first read, recognition and sorting unit 3 performs the operations shown in the FIG. 2 flow chart.
- these comprise a first block 100 for acquiring (in known manner) an image of a mail item 7 fed to unit 3 .
- Block 100 is followed by a block 110 , in which the acquired image is subjected to a first (known) character recognition process, typically an OCR (Optical Character Recognition) process, to automatically recognize the alphanumeric characters (receiver's address, postal code, receiver's identification data) in the mail item image.
- a first character recognition process typically an OCR (Optical Character Recognition) process, to automatically recognize the alphanumeric characters (receiver's address, postal code, receiver's identification data) in the mail item image.
- OCR Optical Character Recognition
- Block 110 is followed by a block 120 , which determines success of the process in block 110 , i.e. whether the address has been recognized automatically from the mail item image.
- the address is recognized using known processes, which compare components of the acquired address with data in predetermined tables.
- block 120 is followed by a block 130 , which selects another image, if there is one (if there is not, the process ends), and then goes back to block 100 .
- a first set of recognized data DATA OCR is generated, is stored in a memory of first unit 3 , and is associated with mail item 7 by an identification code, which is applied physically to the item, e.g. in the form of a bar code.
- the bar code may be present already; in which case, if it identifies the mail item unequivocally, it is read and used.
- block 120 is followed by a block 140 , in which the current image is subjected to a known second character recognition process (different from the first), typically an RCR (Reject Character Recognition) process, to automatically recognize the alphanumeric characters in the mail item image.
- a known second character recognition process typically an RCR (Reject Character Recognition) process
- Block 140 is followed by a block 150 , which (like lock 120 ) determines success of the process in block 40 , i.e. whether an address has been recognized.
- block 150 is followed by block 130 , which selects another image and, as before, then goes back to block 100 .
- a second set of recognized data DATA RCR is generated, is stored in a memory of first unit 3 , and is associated with mail item 7 by an identification code.
- block 150 is followed by a block 160 , in which the current image is subjected to a known third character recognition process, involving operator intervention on a video-terminal (not shown), to recognize the address on the mail item.
- a third set of recognized data DATA V-C is therefore generated, is stored in a memory of first unit 3 , and is associated with mail item 7 by an identification code.
- Block 160 then goes on to block 130 .
- a mail item image is therefore definitely subjected to a first recognition process. If the first process fails, the image is subjected to a second recognition process different from the first. And, if this also fails, the image is subjected to video-coding recognition.
- Standby unit 8 performs the operations shown in the FIG. 3 flow chart, during which, in the example shown, the mail items are located in the standby unit 8 shown in FIG. 1 .
- the operations comprise a first block 200 , which extracts an image successfully processed by the first process (in block 110 ), i.e. an image whose mailing address has been recognized and for which relative data DATA OCR is available.
- Block 200 is followed by a block 210 , which subjects the current image to the second, i.e. RCR, recognition process.
- RCR the second, i.e. RCR
- Block 210 is followed by a block 220 , which determines success of the RCR process and whether the address recognized matches the one recognized in the same image by the OCR process (block 110 ).
- block 220 is followed by a block 230 , which memorizes data DATA RCR , selects another image, and then goes back to block 200 to examine the next image, until all the available images have been examined.
- block 220 is followed by a block 300 (queuing block), which inserts the current image (for which the RCR process and the preceding OCR process in FIG. 2 give different addresses, or for which the FIG. 3 RCR process has failed) in a queue to which all the images not recognized by the OCR process in FIG. 2 (NO output of block 120 ) are sent.
- a block 300 queuing block
- the queue therefore comprises:
- Block 300 is followed by a block 310 , which extracts images from the queue for operator video-terminal checking.
- Images are extracted in such a way as to form homogeneous queues, i.e. containing one city place-name (e.g. a street) and arranged by alphanumerically arrangeable elements, e.g. building numbers.
- the images are displayed on the video-terminal (not shown) so that the addresses can be seen and either accepted or modified by the operator. It should be pointed out that the system provides for displaying the unmatched addresses (e.g. those coded differently by blocks 120 and 210 ), which are conveniently displayed together with the image of the relative mail item.
- the operator may confirm the proposed address, select one of the proposed addresses (in the event of unmatched addresses as described above), alter the proposed address completely, or retrieve the image of a mail item whose address has been altered manually, e.g. to check the proposed address has been altered correctly.
- the operator may:
- block 310 Upon completion of the operations in block 310 , the operator enters an end-of-check command, and block 310 goes on to a block 320 , which stores the addresses DATA CHK checked by the operator in block 310 and then goes back to block 310 until all the images have been checked.
- Second unit 9 performs the operations shown in the FIG. 4 flow chart.
- these comprise a first block 400 , which acquires (in known manner) another image of a mail item 7 fed to unit 9 , acquires the identification code/s previously applied to the mail item, and is followed by a block 410 , in which the acquired image is subjected to the OCR (Optical Character Recognition) process.
- OCR Optical Character Recognition
- Block 410 therefore repeats the process already performed by block 110 .
- Block 410 is followed by a block 420 , which determines success of the process in block 410 , i.e. whether the address has been recognized automatically from the mail item image.
- block 420 In the event of a positive response (address recognized—first process successful), block 420 is followed by a block 430 , which performs a global statistical mail item acceptance analysis by means of tables, selects another image, if there is one, and then goes back to block 400 .
- a set of recognized data DATA OCR-2 is generated, is stored in a memory of first unit 9 , and is associated with mail item 7 using the existing identification code.
- block 420 is followed by a block 440 , which subjects the current image to an RCR (Reject Character Recognition) process.
- RCR Reject Character Recognition
- Block 440 is followed by a block 450 , which stores the previously acquired data DATA RCR-2 .
- Block 450 then goes back to block 430 .
- the second unit repeats the same automatic OCR and RCR operations performed by first unit 3 , but with no video-coding operations.
- the results of the operations performed by the second unit are stored to permit real-time checking by which to sort and statistically control the stored acquired data.
- the object of the method according to the present invention is to minimize error.
- the mail items are rejected or sent for checking to a further video-coding station (not shown).
- the image of a mail item for which a possible double extraction has been determined may be forwarded for operator checking (video-coding).
- the mechanical line forwards the mail item to a video-coding recognition system, by which an operator determines the actual nature of the mail item image in question.
- the real-time check is performed using the tables in block 430 .
- the FIG. 5 table shows an example of one possible check.
- a row R 1 -R 5 and a column C 1 -C 27 are selected, and the intersection of the row and column indicates a box indicating the operation (sort/reject) to be performed on the mail item.
Abstract
Description
- The present invention relates to a mail recognition method.
- Mail recognition methods are known in which images of mail items undergo a first character recognition process to generate a first set of data representing alphanumeric characters (indicating the postal code, receiver's address and data) on the mail items.
- The first recognition process is normally an OCR process, and usually succeeds in recognizing the address on the basis of the alphanumeric characters (roughly 10% of the images fail to be recognized, and the recognition error margin of the OCR process is about 2-3%).
- The images the first recognition process has failed to recognize (i.e. about 10%) undergo a second character recognition process, different from the first, to generate a second set of data, obviously also representing alphanumeric characters on the mail items.
- The second recognition process may, for example, be an RCR (Reject Character Recognition) process, which usually succeeds in recognizing 40% of the preceding OCR process rejects.
- There may also be an operator coding process.
- Cascading the above three processes increases the total mail recognition percentage to as much as over 99%, while at the same time reducing the error percentage to less than 0.5%.
- As is known, recognition algorithms may be variously parameterized to improve recognition probability at the expense of recognition accuracy (probability complementary to error probability per recognized volume).
- In some applications, recognition and accuracy percentages may be insufficient to ensure successful follow-up mail processing, such as highly automated sorting processes.
- It is an object of the present invention to provide a mail recognition process designed to ensure highly accurate recognition combined with a high mail item recognition percentage.
- According to the present invention, there is provided a mail recognition method comprising the steps of: acquiring images of mail items, and subjecting said images to a first character recognition process to recognize addresses; and characterized by comprising the step of subjecting the images successfully processed by said first character recognition process to a second character recognition process, different from the first, to recognize an address; and checking to determine whether the results of the different character recognition processes performed on the same image match.
- A preferred, non-limiting embodiment of the invention will be described by way of example with reference to the accompanying drawings, in which:
-
FIG. 1 shows, schematically, a mail recognition system implementing the method according to the present invention; -
FIG. 2 shows first steps in the method according to the present invention; -
FIG. 3 shows second steps in the method according to the present invention; -
FIG. 4 shows third steps in the method according to the present invention; -
FIG. 5 shows a table used in the method according to the present invention. -
Number 1 inFIG. 1 indicates as a whole a mail processing system implementing the method according to the present invention. - More specifically,
system 1 comprises: -
- a first read, recognition and sorting
unit 3 receiving a stream F of mail items 7 (letters, postcards, enclosed documents, flats, cellophane-wrapped magazines, unwrapped magazines, catalogues, etc.); - a
standby unit 8 receivingmail items 7 from first read, recognition and sortingunit 3; and - a second read, recognition and sorting
unit 9 receivingmail items 7 fromstandby unit 8.
- a first read, recognition and sorting
- The second read, recognition and sorting
unit 9 emits mail items 7 (letters, postcards, enclosed documents, flats, cellophane-wrapped magazines, unwrapped magazines, catalogues, etc.), and possibly also sends them to other units (not shown) for performing known mail processing operations, such as a mail sequencing process. - The first read, recognition and
sorting unit 3 performs the operations shown in theFIG. 2 flow chart. - More specifically, these comprise a
first block 100 for acquiring (in known manner) an image of amail item 7 fed tounit 3. -
Block 100 is followed by ablock 110, in which the acquired image is subjected to a first (known) character recognition process, typically an OCR (Optical Character Recognition) process, to automatically recognize the alphanumeric characters (receiver's address, postal code, receiver's identification data) in the mail item image. -
Block 110 is followed by ablock 120, which determines success of the process inblock 110, i.e. whether the address has been recognized automatically from the mail item image. The address is recognized using known processes, which compare components of the acquired address with data in predetermined tables. - In the event of a positive response (address recognized from the alphanumeric characters and/or bar code (if any) in the image—first process successful),
block 120 is followed by ablock 130, which selects another image, if there is one (if there is not, the process ends), and then goes back toblock 100. - If the address on the mail item is recognized, a first set of recognized data DATAOCR is generated, is stored in a memory of
first unit 3, and is associated withmail item 7 by an identification code, which is applied physically to the item, e.g. in the form of a bar code. Alternatively, the bar code may be present already; in which case, if it identifies the mail item unequivocally, it is read and used. - In the event of a negative response (address on the mail item not recognized, or the image not processed in block 110),
block 120 is followed by ablock 140, in which the current image is subjected to a known second character recognition process (different from the first), typically an RCR (Reject Character Recognition) process, to automatically recognize the alphanumeric characters in the mail item image. -
Block 140 is followed by ablock 150, which (like lock 120) determines success of the process in block 40, i.e. whether an address has been recognized. - In the event of a positive response (address recognized by the second process),
block 150 is followed byblock 130, which selects another image and, as before, then goes back toblock 100. - If the address on the mail item is recognized, a second set of recognized data DATARCR is generated, is stored in a memory of
first unit 3, and is associated withmail item 7 by an identification code. - In the event of a negative response (address on the mail item not recognized, or the image not processed in block 140),
block 150 is followed by ablock 160, in which the current image is subjected to a known third character recognition process, involving operator intervention on a video-terminal (not shown), to recognize the address on the mail item. - A third set of recognized data DATAV-C is therefore generated, is stored in a memory of
first unit 3, and is associated withmail item 7 by an identification code. -
Block 160 then goes on to block 130. - A mail item image is therefore definitely subjected to a first recognition process. If the first process fails, the image is subjected to a second recognition process different from the first. And, if this also fails, the image is subjected to video-coding recognition.
-
Standby unit 8 performs the operations shown in theFIG. 3 flow chart, during which, in the example shown, the mail items are located in thestandby unit 8 shown inFIG. 1 . - More specifically, the operations comprise a
first block 200, which extracts an image successfully processed by the first process (in block 110), i.e. an image whose mailing address has been recognized and for which relative data DATAOCR is available. -
Block 200 is followed by ablock 210, which subjects the current image to the second, i.e. RCR, recognition process. -
Block 210 is followed by ablock 220, which determines success of the RCR process and whether the address recognized matches the one recognized in the same image by the OCR process (block 110). - In the event of a positive response (RCR process successful, and matching addresses recognized by both different processes),
block 220 is followed by ablock 230, which memorizes data DATARCR, selects another image, and then goes back to block 200 to examine the next image, until all the available images have been examined. - In the event of a negative response,
block 220 is followed by a block 300 (queuing block), which inserts the current image (for which the RCR process and the preceding OCR process inFIG. 2 give different addresses, or for which theFIG. 3 RCR process has failed) in a queue to which all the images not recognized by the OCR process inFIG. 2 (NO output of block 120) are sent. - The queue therefore comprises:
-
- images rejected by
block 120 operating as shown in theFIG. 2 flow chart, i.e. the images rejected by the first OCR process on the images, and which were therefore coded by the RCR process (block 140) or video-coded (block 160); - images not recognized by the RCR process in block 210 (
FIG. 3 ); and - images whose addresses recognized in
block 210 do not match those inblock 110.
- images rejected by
-
Block 300 is followed by ablock 310, which extracts images from the queue for operator video-terminal checking. - Images are extracted in such a way as to form homogeneous queues, i.e. containing one city place-name (e.g. a street) and arranged by alphanumerically arrangeable elements, e.g. building numbers.
- The images are displayed on the video-terminal (not shown) so that the addresses can be seen and either accepted or modified by the operator. It should be pointed out that the system provides for displaying the unmatched addresses (e.g. those coded differently by
blocks 120 and 210), which are conveniently displayed together with the image of the relative mail item. - The operator may confirm the proposed address, select one of the proposed addresses (in the event of unmatched addresses as described above), alter the proposed address completely, or retrieve the image of a mail item whose address has been altered manually, e.g. to check the proposed address has been altered correctly.
- In other words, the operator may:
-
- complete the available data by adding, deleting, or altering the stored addresses; and
- select the available data (e.g. in the event the OCR and RCR process results fail to match).
- Upon completion of the operations in
block 310, the operator enters an end-of-check command, andblock 310 goes on to ablock 320, which stores the addresses DATACHK checked by the operator inblock 310 and then goes back toblock 310 until all the images have been checked. -
Second unit 9 performs the operations shown in theFIG. 4 flow chart. - More specifically, these comprise a
first block 400, which acquires (in known manner) another image of amail item 7 fed tounit 9, acquires the identification code/s previously applied to the mail item, and is followed by ablock 410, in which the acquired image is subjected to the OCR (Optical Character Recognition) process. -
Block 410 therefore repeats the process already performed byblock 110. -
Block 410 is followed by ablock 420, which determines success of the process inblock 410, i.e. whether the address has been recognized automatically from the mail item image. - In the event of a positive response (address recognized—first process successful), block 420 is followed by a
block 430, which performs a global statistical mail item acceptance analysis by means of tables, selects another image, if there is one, and then goes back to block 400. - If the address on the mail item is recognized, a set of recognized data DATAOCR-2 is generated, is stored in a memory of
first unit 9, and is associated withmail item 7 using the existing identification code. - In the event of a negative response (mail item address not recognized), block 420 is followed by a
block 440, which subjects the current image to an RCR (Reject Character Recognition) process. -
Block 440 is followed by ablock 450, which stores the previously acquired data DATARCR-2. -
Block 450 then goes back to block 430. - In other words, the second unit repeats the same automatic OCR and RCR operations performed by
first unit 3, but with no video-coding operations. - Like those performed by the first unit, the results of the operations performed by the second unit are stored to permit real-time checking by which to sort and statistically control the stored acquired data.
- The object of the method according to the present invention, in fact, is to minimize error. For which purpose, when statistical analysis reveals potentially conflicting situations (e.g. non-matching recognized addresses or identification bar codes which do not conform with the recognition result), the mail items are rejected or sent for checking to a further video-coding station (not shown).
- In addition, the image of a mail item for which a possible double extraction has been determined (in known manner) may be forwarded for operator checking (video-coding). In which case, when a potential double extraction is determined, the mechanical line forwards the mail item to a video-coding recognition system, by which an operator determines the actual nature of the mail item image in question.
- The real-time check is performed using the tables in
block 430. - The
FIG. 5 table shows an example of one possible check. - More specifically:
-
- row R1 is selected in the event of only one semiautomatic address recognition (i.e. by video-coding, block 160—
FIG. 2 ); - row R2 is selected in the event of only one automatic address recognition (i.e. by the OCR or RCR process in
block 110 or 140); - row R3 is selected in the event of two automatic address recognitions by
blocks - row R4 is selected in the event of two automatic address recognitions by
blocks - row R5 is selected in the event of operator-entered address data in
block 310.
- row R1 is selected in the event of only one semiautomatic address recognition (i.e. by video-coding, block 160—
- Columns C1-C27 in the table are selected on the basis of a matrix M, which receives as input data:
-
- the comparison between the result of the OCR process in
block 410 and the operator-entered result inblock 310; or - the comparison between the result of the RCR process in
block 440 and the operator-entered result inblock 310.
- the comparison between the result of the OCR process in
- On the basis of the data associated with each mail item image, a row R1-R5 and a column C1-C27 are selected, and the intersection of the row and column indicates a box indicating the operation (sort/reject) to be performed on the mail item.
Claims (14)
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
ITTO2004A000299 | 2004-05-07 | ||
IT000299A ITTO20040299A1 (en) | 2004-05-07 | 2004-05-07 | POSTAL RECOGNITION METHOD |
Publications (1)
Publication Number | Publication Date |
---|---|
US20050249376A1 true US20050249376A1 (en) | 2005-11-10 |
Family
ID=34939720
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/122,241 Abandoned US20050249376A1 (en) | 2004-05-07 | 2005-05-05 | Mail recognition method |
Country Status (3)
Country | Link |
---|---|
US (1) | US20050249376A1 (en) |
EP (1) | EP1594077A3 (en) |
IT (1) | ITTO20040299A1 (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080273749A1 (en) * | 2007-05-02 | 2008-11-06 | Lockheed Martin Corporation | Implicit video coding confirmation of automatic address recognition |
US20090219556A1 (en) * | 2008-02-29 | 2009-09-03 | Fuji Xerox Co., Ltd. | Paper document processing apparatus, paper document processing method, and computer readable medium |
US20090285447A1 (en) * | 2008-05-15 | 2009-11-19 | Rundle Alfred T | Correcting video coding errors using an automatic recognition result |
US20090285486A1 (en) * | 2005-08-26 | 2009-11-19 | Siemens Aktiengesellschaft | Method for identifying postal mailings |
US20100232642A1 (en) * | 2007-05-30 | 2010-09-16 | Solystic | Method of Processing Mailpieces, The Method Including Graphically Classifying Signatures Associated With The Mailpieces |
US20140064622A1 (en) * | 2012-09-06 | 2014-03-06 | Xerox Corporation | Method and system for evaluating handwritten documents |
US20140149341A1 (en) * | 2012-11-29 | 2014-05-29 | Electronics And Telecommunications Research Institute | System and method for refining address database for improving performance of automated mail sorting machine |
US20220180669A1 (en) * | 2020-12-03 | 2022-06-09 | Ricoh Company, Ltd. | Data stream enabled return mail services |
EP4120129A4 (en) * | 2020-03-09 | 2024-02-28 | Toshiba Kk | Recognition device and program |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010009410A (en) * | 2008-06-27 | 2010-01-14 | Toshiba Corp | Video coding system, classifying system, coding method and classifying method |
JP2016159271A (en) * | 2015-03-04 | 2016-09-05 | 株式会社東芝 | Delivery processor and delivery processing program |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5889897A (en) * | 1997-04-08 | 1999-03-30 | International Patent Holdings Ltd. | Methodology for OCR error checking through text image regeneration |
US20030132572A1 (en) * | 2001-12-03 | 2003-07-17 | Andre Rompe | Multiples detect apparatus and method |
US20030179906A1 (en) * | 1999-09-24 | 2003-09-25 | Mailcode Inc. A Subsidiary Of Pitney Bowes Inc. | Mail processing systems and methods |
US20030182017A1 (en) * | 2000-09-08 | 2003-09-25 | O'donohue Gerard | Systems and methods for sorting mail using a name/firm database |
US6665422B1 (en) * | 1996-11-12 | 2003-12-16 | Siemens Aktiengesellchaft | Method and device for recognizing distribution data on postal packets |
US6741724B1 (en) * | 2000-03-24 | 2004-05-25 | Siemens Dematic Postal Automation, L.P. | Method and system for form processing |
-
2004
- 2004-05-07 IT IT000299A patent/ITTO20040299A1/en unknown
-
2005
- 2005-05-05 US US11/122,241 patent/US20050249376A1/en not_active Abandoned
- 2005-05-05 EP EP05103782A patent/EP1594077A3/en not_active Ceased
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6665422B1 (en) * | 1996-11-12 | 2003-12-16 | Siemens Aktiengesellchaft | Method and device for recognizing distribution data on postal packets |
US5889897A (en) * | 1997-04-08 | 1999-03-30 | International Patent Holdings Ltd. | Methodology for OCR error checking through text image regeneration |
US20030179906A1 (en) * | 1999-09-24 | 2003-09-25 | Mailcode Inc. A Subsidiary Of Pitney Bowes Inc. | Mail processing systems and methods |
US6741724B1 (en) * | 2000-03-24 | 2004-05-25 | Siemens Dematic Postal Automation, L.P. | Method and system for form processing |
US20030182017A1 (en) * | 2000-09-08 | 2003-09-25 | O'donohue Gerard | Systems and methods for sorting mail using a name/firm database |
US20030132572A1 (en) * | 2001-12-03 | 2003-07-17 | Andre Rompe | Multiples detect apparatus and method |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9323998B2 (en) * | 2005-08-26 | 2016-04-26 | Siemens Aktiengesellschaft | Method for identifying postal mailings |
US20090285486A1 (en) * | 2005-08-26 | 2009-11-19 | Siemens Aktiengesellschaft | Method for identifying postal mailings |
US7925046B2 (en) | 2007-05-02 | 2011-04-12 | Lockheed Martin Corporation | Implicit video coding confirmation of automatic address recognition |
US20080273749A1 (en) * | 2007-05-02 | 2008-11-06 | Lockheed Martin Corporation | Implicit video coding confirmation of automatic address recognition |
US20100232642A1 (en) * | 2007-05-30 | 2010-09-16 | Solystic | Method of Processing Mailpieces, The Method Including Graphically Classifying Signatures Associated With The Mailpieces |
US8126204B2 (en) * | 2007-05-30 | 2012-02-28 | Solystic | Method of processing mailpieces, the method including graphically classifying signatures associated with the mailpieces |
US8339623B2 (en) * | 2008-02-29 | 2012-12-25 | Fuji Xerox Co., Ltd. | Paper document processing apparatus, paper document processing method, and computer readable medium |
US20090219556A1 (en) * | 2008-02-29 | 2009-09-03 | Fuji Xerox Co., Ltd. | Paper document processing apparatus, paper document processing method, and computer readable medium |
US8005260B2 (en) * | 2008-05-15 | 2011-08-23 | Graham Packaging Company L.P. | Correcting video coding errors using an automatic recognition result |
US20090285447A1 (en) * | 2008-05-15 | 2009-11-19 | Rundle Alfred T | Correcting video coding errors using an automatic recognition result |
US20140064622A1 (en) * | 2012-09-06 | 2014-03-06 | Xerox Corporation | Method and system for evaluating handwritten documents |
US9098777B2 (en) * | 2012-09-06 | 2015-08-04 | Xerox Corporation | Method and system for evaluating handwritten documents |
US20140149341A1 (en) * | 2012-11-29 | 2014-05-29 | Electronics And Telecommunications Research Institute | System and method for refining address database for improving performance of automated mail sorting machine |
EP4120129A4 (en) * | 2020-03-09 | 2024-02-28 | Toshiba Kk | Recognition device and program |
US20220180669A1 (en) * | 2020-12-03 | 2022-06-09 | Ricoh Company, Ltd. | Data stream enabled return mail services |
Also Published As
Publication number | Publication date |
---|---|
EP1594077A3 (en) | 2006-08-09 |
EP1594077A2 (en) | 2005-11-09 |
ITTO20040299A1 (en) | 2004-08-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20050249376A1 (en) | Mail recognition method | |
JP4580233B2 (en) | Mail identification tag with image signature and associated mail handler | |
US7539326B2 (en) | Method for verifying an intended address by OCR percentage address matching | |
US6909789B1 (en) | Method of processing postal matters | |
JPH0694026B2 (en) | Delivery processing method and system | |
KR100323351B1 (en) | Address recognizing method and mail processing apparatus | |
KR100536509B1 (en) | Method and device for recognition of delivery data on mail matter | |
US7181045B2 (en) | Method and device for reading the addresses of items of mail | |
EP0054842A2 (en) | Figure recognizing system | |
US7219095B1 (en) | Installation and method for updating an address database with recorded address records | |
US5852685A (en) | Enhanced batched character image processing | |
US5526447A (en) | Batched character image processing | |
JP5456776B2 (en) | How to classify multiple objects in data analysis | |
JPH0739820A (en) | Street zone recognizing device and address reading and classifying machine | |
US7215794B2 (en) | Method and device for reading the addresses on items of mail | |
JPH08101879A (en) | Mail processor | |
US10471477B2 (en) | Method of sorting pre-sorted mailpieces | |
JP3710866B2 (en) | Mail sorting apparatus, mail processing system, and mail processing method | |
US8126917B2 (en) | Method and device for transporting an object to a destination address specified incompletely | |
US6993155B1 (en) | Method for reading document entries and addresses | |
US20090161910A1 (en) | Method and Device for Reading Addresses of Mail Items | |
JP2002056344A (en) | Information processing device, information processing method, paper sheet partitioning device, and paper sheet partitioning method | |
JPH0957199A (en) | Address reading apparatus and mail-sorting apparatus | |
JPH11179289A (en) | Mail classifier | |
JP2021121892A (en) | Number plate recognition device, number plate recognition method, and number plate recognition program |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: ELSAG SPA, ITALY Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:DE LEO, GUIDO;FRANZONE, CRISTIANO;REEL/FRAME:016181/0582 Effective date: 20050616 |
|
AS | Assignment |
Owner name: ELSAG SPA, ITALY Free format text: CORRECTIVE ASSIGNMENT TO CORRECT THE A THIRD INVENTOR TO BE ADDED PREVIOUSLY RECORDED ON REEL 016181 FRAME 0582;ASSIGNORS:DE LEO, GUIDO;FRANZONE, CRISTIANO;VICENZI, MICHELE;REEL/FRAME:016733/0302 Effective date: 20050616 |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION |